Secondhand Smoke Exposure and Inflammatory Markers in ...

3 downloads 11540 Views 263KB Size Report
opera tors, freight handlers, mechanics), it remains ... Medical Research Institute (FAMRI; F.L.);. National ...... as diesel exhaust, may confound or modify the.
Research Secondhand Smoke Exposure and Inflammatory Markers in Nonsmokers in the Trucking Industry Yueh-Hsiu Mathilda Chiu,1,2 Donna Spiegelman,3,4 Douglas W. Dockery,1,3 Eric Garshick,2,5 S. Katharine Hammond,6 Thomas J. Smith,1 Jaime E. Hart,2,3 and Francine Laden1,2,3 1Exposure,

Epidemiology and Risk Program, Department of Environmental Health, Harvard School of Public Health, Boston, MA, USA; Laboratory, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts, USA; 3Department of Epidemiology, and 4Department of Biostatistics, Harvard School of Public Health, Boston, Massachusetts, USA; 5Pulmonary and Critical Care Medicine Section, VA Boston Healthcare System, Boston, Massachusetts, USA; 6School of Public Health, University of California, Berkeley, California, USA 2Channing

Background: Few studies have directly assessed the association of secondhand smoke (SHS) with cardiovascular disease–related inflammatory markers, and the findings are inconsistent. Objectives: We assessed the association between SHS exposure and the inflammatory markers high-sensitivity C-reactive protein (hs‑CRP), interleukin-6 (IL‑6), and soluble inter­cellular adhesion molecule-1 (sICAM‑1) in 199 non­smoking U.S. trucking industry workers. Methods: Participants provided blood samples either by mail (blood drawn at local health care provider near home) or at the work site (blood drawn by research staff on-site) and completed a health and work history questionnaire at the time of blood draw. Exposure to SHS was measured by plasma cotinine concentrations. We used multivariate regression analyses to assess the associations between levels of cotinine and inflammatory markers. Results: The median cotinine level was 0.10 ng/mL (interquartile range, 0.04–0.23 ng/mL). The odds ratios of elevated hs‑CRP (above highest CRP tertile, 1.5 mg/L) were 2.85 [95% confidence interval (CI), 1.03–7.89] for the high-cotinine group (> 0.215 ng/mL) and 2.80 (95% CI, 1.11–7.10) for the moderate-cotinine group (0.05–0.215 ng/mL), compared with the low-cotinine group ( LOQ. Samples  14 ng/mL is generally considered to be a smoker (Jarvis et al. 1987). Of 218 workers who reported no current cigarette, cigar, or pipe smoking, 199 (91%; 94  recruited by mail and 105  on-site) had cotinine levels ≤���������������������������������������������  14 ng/mL and were eligible for our statistical analysis. None of the participants reported taking nicotine replacement therapy. Assessment of inflammatory markers. The analyses of inflammatory markers were performed at the Clinical and Epidemiologic Research Laboratory, Department of Laboratory Medicine, at Children’s Hospital (Boston, MA). CRP levels were determined by an immuno­t urbidi­m etric assay on a Hitachi  917 analyzer (Roche Diagnostics, Indianapolis, IN), using reagents and calibrators from Denka Seiken (Niigata, Japan). Concentrations of sICAM‑1 and IL‑6 were measured by enzyme-linked immunosorbent assays (ELISA), using the quantitative sandwich enzyme immuno­assay technique (R&D Systems, Minneapolis, MN). For the three inflammatory markers analyzed, all the samples were above the laboratory thresholds of detection (0.03 mg/L for CRP, 0.094 pg/mL for IL‑6, and 0.35 ng/mL for sICAM‑1). The day-to-day variability of the analytic assay was 2.81% for CRP, 7.2% for IL‑6, and 6.0% for sICAM‑1. The Spearman’s correlation coefficient was 0.43 (p  10 mg/L. The results were consistent with analyses for the population as a whole.

Discussion In this population of non­smoking blue-collar trucking industry workers, we observed evidence of an association between plasma cotinine levels and elevated hs‑CRP, but not IL‑6 or sICAM‑1. Self-reported SHS exposure in the past 24 hr was significantly positively associated with increased level of plasma cotinine, indicating that self-reported SHS exposure was reliable in this population. A few previous studies have examined the association between SHS exposure and CRP, with inconsistent results. In studies of NHANES III, no significant association was found between CRP and serum cotinine levels in never-smokers in the U.S. general population (Venn and Britton 2007) or non­ smoking adult workers without home SHS exposure (Clark et al. 2008). On the other hand, in an analysis of Greek adults who had never smoked (n = 995), Panagiotakos et al. (2004) reported a 0.8 mg/L (p = 0.03) increase of CRP for those regularly exposed to SHS (> 3 days/week for at least 30 min/ day), compared with those who were not exposed. In another study, Jefferis et al. (2010) investigated about 5,000 non­smoking British elderly and observed a significant association between serum cotinine and CRP in a linear regression model adjusted for age, sex, region of residence, health behaviors, social class, and BMI (β = 0.03; 95% CI, 0.01–0.05; difference of natural log–­transformed CRP associated with a doubling in cotinine concentration). The results from the present study suggest that higher plasma cotinine levels (> 0.05 ng/mL) may be associated with elevated CRP (> 1.5 mg/L). The analytic method for CRP used in our study is a high-sensitivity

immuno­a ssay with a detection limit of 0.03 mg/L, compared with 3 mg/L (Venn and Britton 2007) and 0.1 mg/L (Clark et al. 2008) in the NHANES III studies. To our knowledge, the association between SHS exposure and IL‑6 was investigated previously only in a study of British elderly (Jefferis et al. 2010). In multivariateadjusted linear regression, that study showed a marginally significant positive association between log-transformed IL‑6 and cotinine (β  =  0.01; 95% CI, 0.00–0.02; associated with a doubling in cotinine concentration),

whereas our study did not show this association. No previous studies have estimated the effect of SHS exposure on sICAM‑1, and we did not observe an association in the present study. Other markers of inflammation, such as homo­cysteine and fibrino­ gen, have been associated with serum cotinine levels (Clark et  al. 2008; Jefferis et al. 2010; Panagiotakos et al. 2004; Venn and Britton 2007). It is possible that the bio­markers of inflammation that we examined are less sensitive to low level SHS exposure than are homocysteine and fibrinogen.

Table 2. Levels of inflammatory markers [median (IQR)] by demographic, work-related, and health-related variables. Variable Age (years) ≤ 50 50–60 > 60 Sex Female Male Race Non-Caucasian Caucasian Education (years) ≤ 12 > 12 BMIa Normal Overweight Obese Smoking history Never Former Alcohol use in previous 24 hr No Yes Job title Office worker Terminal worker Driver Retired Recruitment group On-site By mail COPD No Yes Asthma No Yes Chronic rhinitis No Yes High blood pressure No Yes Heart problem/disease No Yes Hay fever No Yes Aspirin use No Yes aNormal,

n

hs-CRP (mg/L)

IL-6 (pg/mL)

sICAM-1 (ng/mL)

57 76 66

0.83 (0.43–1.90) 0.84 (0.44–1.84) 1.14 (0.63–2.45)

1.22 (0.84–2.05) 1.41 (1.08–2.42) 1.64 (1.05–3.39)

239.7 (202.3–282.8) 245.4 (198.7–297.0) 252.2 (214.8–283.4)

16 183

1.48 (0.86–3.96) 0.91 (0.44–1.92)

1.39 (0.99–2.48) 1.38 (1.02–2.66)

277.5 (237.8–328.2) 242.4 (201.3–284.3)

16 183

0.99 (0.47–3.41) 0.93 (0.46–1.96)

1.32 (0.83–1.83) 1.38 (1.04–2.66)

260.0 (190.9–297.0) 246.0 (203.7–284.3)

128 71

0.98 (0.46–2.13) 0.83 (0.46–1.72)

1.32 (0.93–2.73) 1.44 (1.07–2.25)

242.0 (200.3–281.8) 254.2 (208.0–291.6)

27 95 77

0.71 (0.42–1.13) 0.73 (0.42–1.42) 1.91 (0.84–3.65)

1.31 (0.87–1.98) 1.18 (0.88–2.05) 1.86 (1.16–3.60)

241.6 (230.9–276.6) 239.4 (199.4–279.3) 260.9 (208.1–301.1)

101 98

0.76 (0.42–1.69) 1.28 (0.68–2.52)

1.21 (0.88–2.05) 1.73 (1.11–3.10)

239.7 (199.0–276.6) 255.8 (207.8–299.8)

136 43

0.98 (0.50–2.08) 0.73 (0.32–1.73)

1.44 (1.10–2.69) 1.38 (0.84–2.70)

247.7 (207.6–284.4) 238.9 (197.8–283.4)

18 58 97 26

1.00 (0.82–1.96) 0.87 (0.46–1.64) 0.91 (0.42–1.97) 1.14 (0.72–2.57)

1.31 (1.05–1.86) 1.18 (0.88–2.10) 1.42 (1.03–2.99) 1.74 (1.18–3.52)

282.3 (239.0–321.7) 234.6 (203.6–268.0) 244.6 (198.4–283.6) 252.4 (214.8–306.6)

105 94

0.92 (0.43–2.09) 0.95 (0.55–1.96)

1.42 (1.03–2.66) 1.24 (1.01–2.57)

233.2 (197.8–273.5) 256.7 (217.7–299.8)

183 16

0.86 (0.44–1.88) 2.49 (1.41–4.27)

1.34 (0.96–2.48) 1.81 (1.18–4.25)

241.6 (202.0–282.8) 319.3 (266.7–344.0)

179 20

0.91 (0.44–1.91) 1.89 (0.79–2.37)

1.34 (1.02–2.47) 1.95 (1.17–4.13)

242.8 (202.3–282.8) 283.9 (222.0–339.6)

184 15

0.92 (0.45–1.95) 1.86 (0.71–2.17)

1.32 (0.98–2.48) 2.15 (1.65–2.91)

242.0 (203.1–284.0) 274.3 (242.8–316.9)

140 59

0.82 (0.44–1.88) 1.38 (0.71–2.68)

1.22 (0.88–2.07) 2.05 (1.16–3.34)

242.6 (202.6–284.0) 256.6 (217.2–314.1)

175 24

0.93 (0.46–1.96) 1.18 (0.54–2.41)

1.32 (1.02–2.48) 2.07 (1.08–4.62)

241.6 (202.3–281.2) 287.0 (210.6–358.9)

156 43

0.90 (0.45–1.90) 1.22 (0.50–2.16)

1.36 (1.03–2.65) 1.40 (0.88–2.48)

246.8 (205.1–289.9) 231.2 (202.0–282.8)

137 62

0.93 (0.46–1.91) 0.89 (0.46–2.37)

1.34 (1.04–2.48) 1.41 (0.92–2.91)

242.4 (200.7–282.8) 254.2 (208.0–290.2)

 median vs. ≤ median)b Truck industry Below LOQ,  0.11 ng/mL NHANES III Low,  0.215 ng/mL ROC curve Low, ≤ 0.06 ng/mL High, > 0.06 ng/mL hs-CRP (highest tertile vs. others)c Truck industry Below LOQ,  0.11 ng/mL NHANES III Low,  0.215 ng/mL ROC curve Low, ≤ 0.06 ng/mL High, > 0.06 ng/mL IL-6 (> median vs. ≤ median)d Truck industry Below LOQ,  0.11 ng/mL NHANES III Low,  0.215 ng/mL ROC curve Low, ≤ 0.06 ng/mL High, > 0.06 ng/mL sICAM-1 (> median vs. ≤ median)e Truck industry Below LOQ,  0.11 ng/mL NHANES III Low,  0.215 ng/mL ROC curve Low, ≤ 0.06 ng/mL High, > 0.06 ng/mL

n

Multivariate adjusteda OR (95% CI) p-Value for trend

19 90 90

Reference 1.05 (0.32–3.48) 1.12 (0.33–3.84)

0.84

65 79 55

Reference 1.78 (0.77–4.09) 1.43 (0.57–3.60)

0.80

73 126

Reference 1.99 (0.92–4.30)

0.08

19 90 90

Reference 1.18 (0.31–4.59) 1.88 (0.48–7.35)

0.21

65 79 55

Reference 2.80 (1.11–7.10) 2.85 (1.03–7.89)

0.17

73 126

Reference 2.72 (1.16–6.37)

0.02

19 90 90

Reference 1.12 (0.35–3.65) 0.94 (0.28–3.14)

0.70

65 79 55

Reference 1.60 (0.69–3.71) 1.21 (0.49–2.99)

0.94

73 126

Reference 1.70 (0.78–3.70)

0.18  

19 90 90

Reference 0.48 (0.15–1.53) 0.70 (0.21–2.26)

0.61

65 79 55

Reference 1.32 (0.60–2.91) 0.66 (0.28–1.55)

0.15

73 126

Reference 1.10 (0.54–2.22)

0.80

Exposure cut-points for NHANES III are from Venn and Britton (2007); those for the ROC curve were calculated following the method of Fawcett (2006). aLogistic regression model adjusted for age, sex, race, educational level, job title, recruitment method, former smoking, obesity, high blood pressure, history of heart disease, COPD, chronic rhinitis, and hay fever. bhs-CRP median, 0.93 mg/L. chs-CRP highest tertile, > 1.5 mg/L. dIL-6 median, 1.38 pg/mL. esICAM-1 median, 246.1 ng/mL.

1298

volume

smokers might be mis­classified as non­smokers and included in the study population. We used 14 ng/mL as the cut-point of plasma cotinine levels for distinguishing smokers from non­smokers, as suggested previously by Jarvis et al. (1987). Recently, Benowitz et al. (2009) suggested that this definition might over­ estimate the number of non­smokers and that the cut-point should be lowered and be specific to sex and race/ethnicity. They proposed an optimal cut-point of 6.79 ng/mL for adult non-Hispanic white males, most of our study population. In the present study, after excluding workers with cotinine levels > 14 ng/mL, the maximum level of plasma cotinine measured in our study population was 4.44 ng/mL, which is well below this optimal cut-point. Thus, those defined as non­smokers in our study would still have been identified as non­ smokers even with this lower cut-point. To our knowledge, there is no well-established cut-point of cotinine for high/heavy SHS exposure versus low/light SHS exposure. Previous studies have used median levels or levels of detection as cutoffs (Clark et al. 2008; Venn and Britton 2007), which are specific to the study population and the laboratory analytic methods. In our study, we explored the association between plasma cotinine and inflammatory markers using several different cotinine cut-points, including the LOQ (0.02 ng/mL), the median value, and an optimal cutoff chosen from a ROC curve based on data from our population, and cut-points used in previous studies of NHANES III data (Clark et al. 2008; Venn and Britton 2007). Although the ORs of elevated hs‑CRP were positive when 0.02 ng/mL was used as the cutpoint for the low cotinine group, this effect estimate was not statistically significant, probably because this cut-point is too low and the group size is too small. The ROC cut-points and the NHANES cut-points seemed to produce consistent results, possibly at least in part because of the similar cut-points for the reference exposure group. Although exploring different cutoffs could increase the possibility of observing an association due to chance, we found a consistent pattern of a positive association between cotinine and CRP levels across all the analyses regardless of the cutoffs. Previous studies have derived some clinical cut-points for CRP. It has been suggested that a person is at low risk of CVD if CRP is  3 mg/L, and these cut-points approximately correspond to tertiles of CRP levels in the general population of U.S. adults (Pearson et al. 2003; Tousoulis et al. 2007). In addition, a meta-analysis of a list of prospective studies showed that subjects in the highest tertile (study-specific cutpoints) of hs‑CRP in general had twice the relative risk of major coronary events than

119 | number 9 | September 2011  •  Environmental Health Perspectives

SHS and inflammatory markers in nonsmoking truckers

those in the lowest tertile (Danesh et al. 1998, 2000). Thus, the CDC/AHA suggested that the measurement of CRP in the adult population be stratified into three tertiles (Pearson et al. 2003; Tousoulis et al. 2007). However, there are no well-established cut-points for sICAM‑1 or IL‑6. In the present study, the median of hs‑CRP was 0.93  mg/L, which was close to 1 mg/L, the clinical cut-point of average risk of CVD versus low risk of CVD. To be consistent, we also used the median as a cut-point for sICAM‑1 and IL‑6 in our analyses. In addition, we explored different clinical cut-points of hs‑CRP suggested by the studies mentioned above. We found a stronger association between SHS exposure and hs‑CRP when we compared the highest tertile of hs‑CRP with the lower two tertiles. Our study has several potential limitations. The measure­ment of plasma cotinine generally reflects the SHS exposure in the previous 1–2 days and might not represent more acute or more chronic exposures if the exposure is largely variable. Previous studies have suggested that IL‑6 levels have a large daily variability (Tousoulis et al. 2007), whereas sICAM‑1 and CRP do not (Mosevold and Bruserud 2002; Osmancik et  al. 2004; Tousoulis et al. 2007). It is possible that we did not completely account for the circadian variation, but the results of our analyses were unchanged when the time of blood draw (i.e., morning, afternoon, evening, overnight) was included in our regression models (data not shown). Our data came from two recruitment sources. Members of the mailing group had their blood samples collected independently using protocols developed for large epidemiologic studies (e.g., the Nurses’ Health Study) and returned them by express mail (Pischon et al. 2003; Ridker et al. 1997). The blood from the on-site group was collected by our study team and similarly express mailed to our laboratory for processing. Despite these collection differences, the distributions of inflammatory markers were similar between the two recruitment groups (Table 2). We also controlled for participant characteristics and the recruitment method in all of the multi­variate regression analyses. In addition, the results from the sensitivity analyses restricted to the on-site group were similar to the findings of the analyses performed for all participants. The range of plasma cotinine and some covariates (e.g., BMI, age) we observed in this study was relatively narrow compared with the NHANES III studies, so we might have missed the association between SHS exposure and inflammatory markers because of non­ differential mis­classifi­ca­tion. Nevertheless, we used high-sensitivity laboratory analytic methods for both plasma cotinine and inflammatory markers, minimizing the chance of mis­ classification for both exposure and outcome.

The sample size of our study was small, leading to the possibility of insufficient power to detect some potentially significant differences. Moreover, other occupational exposures, such as diesel exhaust, may confound or modify the association of SHS exposure with the inflammatory markers. However, we adjusted for job title, a reasonable indicator of occupational exposure to air pollutants in these trucking companies (Smith et al. 2006). Measurements of some conventional biomarkers related to coronary heart disease, such as lipid levels (low-density lipoprotein, high-density lipoprotein, and triglycerides), were not available for our study. However, these are unlikely to confound the association between SHS and inflammatory markers because it is unlikely that these factors would be an antecedent or a proxy marker of an antecedent of SHS. Although information on a history of heart problems and high blood pressure was based on self-report, as opposed to medical records or direct measurements, workers in these trucking companies are required to pass a physical examination conducted every 1–2 years in order to continue working, and health conditions such as high blood pressure and any heart diseases have to be treated. Some retired workers were recruited by mail, but they were all recently retired (within 5 years of the study period). Finally, socio­economic charac­teristics might cause residual confounding. However, our study population is a group of blue-collar workers with fairly homogeneous backgrounds, and we adjusted for years of education. Thus, social factors are unlikely to affect the results we observed in this study.

Conclusion We observed a positive relationship between SHS exposure, as measured by cotinine levels in plasma, and hs‑CRP among non­smoking workers in the trucking industry, although the strength of this association depended on the cut-points selected for analysis. We found no association of plasma cotinine with sICAM‑1 or IL‑6. These results may reflect the relative half-lives of the measured biomarkers and may depend on the methods used to define cutoffs of exposures and health effects. References Bang KM, Kim JH. 2001. Prevalence of cigarette smoking by occupation and industry in the United States. Am J Ind Med 40(3):233–239. Barnoya J, Glantz SA. 2005. Cardiovascular effects of second­ hand smoke: nearly as large as smoking. Circulation 111(20):2684–2698. Benowitz NL. 1996. Cotinine as a biomarker of environmental tobacco smoke exposure. Epidemiol Rev 18(2):188–204. Benowitz NL. 1999. Biomarkers of environmental tobacco smoke exposure. Environ Health Perspect 107(suppl 2):349–355. Benowitz NL, Bernert JT, Caraballo RS, Holiday DB, Wang J. 2009. Optimal serum cotinine levels for distinguishing cigarette smokers and non­smokers within different racial/ ethnic groups in the United States between 1999 and 2004. Am J Epidemiol 169(2):236–248.

Environmental Health Perspectives  •  volume 119 | number 9 | September 2011

Bernert JT Jr, Turner WE, Pirkle JL, Sosnoff CS, Akins JR, Waldrep MK, et al. 1997. Development and validation of sensitive method for determination of serum cotinine in smokers and nonsmokers by liquid chromatography/atmospheric pressure ionization tandem mass spectrometry. Clin Chem 43(12):2281–2291. Blake GJ, Ridker PM. 2002. Inflammatory bio-markers and cardio­ vascular risk prediction. J Intern Med 252(4):283–294. CDC (Centers for Disease Control and Prevention). 2000. Reducing Tobacco Use: A Report of the Surgeon General. Atlanta, GA:CDC. CDC (Centers for Disease Control and Prevention). 2005. Cigarette smoking among adults—United States, 2004. MMWR Morb Mortal Wkly Rep 54(44):1121–1124. CDC (Centers for Disease Control and Prevention). 2006. The Health Consequences of Involuntary Exposure to Tobacco Smoke: A Report of the Surgeon General. Atlanta, GA:CDC. Clark JD III, Wilkinson JD, LeBlanc WG, Dietz NA, Arheart KL, Fleming LE, et al. 2008. Inflammatory markers and secondhand tobacco smoke exposure among U.S. workers. Am J Ind Med 51(8):626–632. Cook DG, Mendall MA, Whincup PH, Carey IM, Ballam L, Morris JE, et al. 2000. C-reactive protein concentration in children: relationship to adiposity and other cardiovascular risk factors. Atherosclerosis 149(1):139–150. Danesh J, Collins R, Appleby P, Peto R. 1998. Association of fibrinogen, C-reactive protein, albumin, or leukocyte count with coronary heart disease: meta-analyses of prospective studies. JAMA 279(18):1477–1482. Danesh J, Whincup P, Walker M, Lennon L, Thomson A, Appleby P, et al. 2000. Low grade inflammation and coronary heart disease: prospective study and updated metaanalyses. BMJ 321(7255):199–204. Eisner MD, Iribarren C. 2007. The influence of cigarette smoking on adult asthma outcomes. Nicotine Tob Res 9(1):53–56. Fawcett T. 2006. An introduction to ROC analysis. Pattern Recognit Lett 27(8):861–874. Gerlach KK, Shopland DR, Hartman AM, Gibson JT, Pechacek TF. 1997. Workplace smoking policies in the United States: results from a national survey of more than 100,000 workers. Tob Control 6(3):199–206. Ho SY, Lam TH, Chung SF, Lam TP. 2007. Cross-sectional and prospective associations between passive smoking and respiratory symptoms at the workplace. Ann Epidemiol 17(2):126–131. Jarvis MJ, Tunstall-Pedoe H, Feyerabend C, Vesey C, Saloojee Y. 1987. Comparison of tests used to distinguish smokers from nonsmokers. Am J Public Health 77(11):1435–1438. Jefferis BJ, Lowe GD, Welsh P, Rumley A, Lawlor DA, Ebrahim S, et al. 2010. Secondhand smoke (SHS) exposure is associated with circulating markers of inflammation and endothelial function in adult men and women. Atherosclerosis 208(2):550–556. Koenig W, Sund M, Frohlich M, Lowel H, Hutchinson WL, Pepys MB. 2003. Refinement of the association of serum C‑reactive protein concentration and coronary heart disease risk by correction for within-subject variation over time: the MONICA Augsburg studies, 1984 and 1987. Am J Epidemiol 158(4):357–364. Lee DJ, Fleming LE, Arheart KL, LeBlanc WG, Caban AJ, ChungBridges K, et al. 2007. Smoking rate trends in U.S. occupational groups: the 1987 to 2004 National Health Interview Survey. J Occup Environ Med 49(1):75–81. Melbye H, Halvorsen DS, Hartz I, Medbo A, Brox J, Eggen AE, et al. 2007. Bronchial airflow limitation, smoking, body mass index, and statin use are strongly associated with the C-reactive protein level in the elderly. The Tromso Study 2001. Respir Med 101(12):2541–2549. Mosevold KA, Bruserud O. 2002. Circadian variations do not have a major impact on serum levels of soluble (S) L-selectin (CD63L), S-intercellular adhesion molecule 1 (sICAM‑1/CD54) and thrombopoietin in healthy individuals. Scand J Clin Lab Invest 62(1):75–79. Nagel G, Arnold FJ, Wilhelm M, Link B, Zoellner I, Koenig W. 2009. Environmental tobacco smoke and cardiometabolic risk in young children: results from a survey in south-west Germany. Eur Heart J 30(15):1885–1893. Osmancik P, Kvasnicka J, Widimsky P, Tarnok A. 2004. Diurnal variation of soluble E- and P-selectin, and intercellular adhesion molecule-1 in patients with and without coronary artery disease. Cardiology 102(4):194–199. Panagiotakos DB, Chrysohoou C, Pitsavos C, Papaioannou I, Skoumas J, Stefanadis C, et  al. 2002. The association between secondhand smoke and the risk of developing

1299

Chiu et al.

acute coronary syndromes, among non-smokers, under the presence of several cardiovascular risk factors: the CARDIO2000 case-control study. BMC Public Health 2:9; doi: 10.1186/1471-2458-2-9 [Online 24 May 2002]. Panagiotakos DB, Pitsavos C, Chrysohoou C, Skoumas J, Masoura C, Toutouzas P, et al. 2004. Effect of exposure to secondhand smoke on markers of inflammation: the ATTICA study. Am J Med 116(3):145–150. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO III, Criqui M, et al. 2003. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation 107(3):499–511. Pischon T, Hankinson SE, Hotamisligil GS, Rifai N, Willett WC, Rimm EB. 2003. Habitual dietary intake of n-3 and n-6 fatty acids in relation to inflammatory markers among US men and women. Circulation 108(2):155–160. Raupach T, Schafer K, Konstantinides S, Andreas S. 2006. Secondhand smoke as an acute threat for the cardiovascular system: a change in paradigm. Eur Heart J 27(4):386–392.

1300

Riches P, Gooding R, Millar BC, Rowbottom AW. 1992. Influence of collection and separation of blood samples on plasma IL-1, IL‑6 and TNF-α concentrations. J Immunol Methods 153(1–2):125–131. Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. 1997. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med 336(14):973–979. Rückerl R, Greven S, Ljungman P, Aalto P, Antoniades C, Bellander T, et al. 2007. Air pollution and inflammation (interleukin-6, C-reactive protein, fibrinogen) in myocardial infarction survivors. Environ Health Perspect 115:1072–1080. Saadeddin SM, Habbab MA, Ferns GA. 2002. Markers of inflammation and coronary artery disease. Med Sci Monit 8(1):RA5–RA12. Saito M, Ishimitsu T, Minami J, Ono H, Ohrui M, Matsuoka H. 2003. Relations of plasma high-sensitivity C-reactive protein to traditional cardiovascular risk factors. Atherosclerosis 167(1):73–79. Smith TJ, Davis ME, Reaser P, Natkin J, Hart JE, Laden F, et al. 2006. Overview of particulate exposures in the US trucking industry. J Environ Monit 8(7):711–720.

volume

Taylor AE, Johnson DC, Kazemi H. 1992. Environmental tobacco smoke and cardiovascular disease. A position paper from the Council on Cardiopulmonary and Critical Care, American Heart Association. Circulation 86(2):699–702. Tousoulis D, Antoniades C, Stefanadis C. 2007. Assessing inflammatory status in cardiovascular disease. Heart 93(8):1001–1007. Venn A, Britton J. 2007. Exposure to secondhand smoke and biomarkers of cardiovascular disease risk in never-smoking adults. Circulation 115(8):990–995. Wannamethee SG, Lowe GD, Shaper AG, Rumley A, Lennon L, Whincup PH. 2005. Associations between cigarette smoking, pipe/cigar smoking, and smoking cessation, and haemostatic and inflammatory markers for cardiovascular disease. Eur Heart J 26(17):1765–1773. White H. 1980. A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica 48(4):817–838. Wortley PM, Caraballo RS, Pederson LL, Pechacek TF. 2002. Exposure to secondhand smoke in the workplace: serum cotinine by occupation. J Occup Environ Med 44(6):503–509.

119 | number 9 | September 2011  •  Environmental Health Perspectives